Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

The text messages pinged with insistent regularity. Liam ignored them. He sat at the bar, nursing a whiskey, ignoring the insistent buzzing of his phone. He’d seen the photo. He knew.

The dim lighting of the bar did nothing to soothe his simmering irritation. He felt it in the clenched muscles of his jaw, in the way his knuckles were white as he gripped the glass. He wasn't ready to face the music. He was the one who'd been wronged, after all.

A woman walked up to the bar, her expression a mix of nervousness and… something else. Recognition? He raised an eyebrow. He did not welcome company, but he wouldn't be the first to leave. He’d ordered his drink first. He would finish it. He wasn't going anywhere.

Emotion: stubborn

Cluster: Defiance / Spite
PC1 (Valence): 0.80 Positive
PC2 (Disposition): 3.73

Role in Research

This story is one of 1,000 stories generated for the emotion stubborn. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 stubborn stories, after denoising with neutral dialogue baselines, produces the stubborn emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the stubborn vector is projected through the unembedding matrix.

Promoted:
de0.780
l0.685
la0.605
either0.580
a0.501
Suppressed:
难以-0.435
终于-0.419
顿时-0.405
竟然-0.396
wondering-0.391